congratulate you, young gentlemen of the class of '87, for the
success you have thus far obtained, and trust that you will persevere in
well doing and win greater success in the future. We need hardly state
that all that has been said was in a spirit of kindness, and we feel
assured that much of it has been seconded by your parents, to whom no
less than to all parents here present off or on the stage, the speaker not
excepted, a serious, thoughtful problem has been, still is, and will
continue to be to many, "What shall we do with our boys."--_Stevens
Indicator._
* * * * *
HEATING MARINE BOILERS WITH LIQUID FUEL.
We were recently witness of an experiment made at Eragny Conflans
on the steam yacht Flamboyante. It was a question of testing a new
vaporizer or burner for liquid fuel. The experiment was a repetition of
the one that the inventor, Mr. G. Dietrich, recently performed with
success in the presence of Admirals Cloue and Miot.
The Flamboyante is 58 ft. in length, 9 ft. in width, draws 5 ft. of water,
and has a displacement of 10 tons. She is provided with a double
vertical engine supplied by a Belleville boiler that develops 28 horse
power. The screw makes 200 revolutions per minute, and gives the
yacht a speed of 6½ knots.
Mr. Dietrich's vaporizer appears to be very simple, and has given so
good results that we have thought it of interest to give our readers a
succinct description of it. In this apparatus, the inventor has endeavored
to obtain an easy regulation of the two essential elements--naphtha and
steam.
Fig. 1 represents the apparatus in section. The steam enters through the
tubulure, A, and finds its way around the periphery of a tuyere, D. It
escapes with great velocity, carries along the petroleum that runs from
two lateral tubulures, B (Fig. 2), and throws it in a fine spray into the
fireplace, through the nozzle, C (Fig. 1), which is flattened into the
shape of a fan opened out horizontally. The mixture at once ignites in
contact with the hot gases, and gives a beautiful, long, clear flame. The
air necessary for the combustion is sucked through the interior of the
nozzle, H, which is in front of the tuyere. It will be seen that the current
of steam can be regulated by moving the tuyere, D, from or toward the
eduction orifice. This is effected through a maneuver of the hand wheel,
F. In the second place, the flow of the petroleum is made regular by
revolving the hand wheel, G, which gives the piston, O, a to and fro
motion in the tuyere, D.
[Illustration: FIG. 1--THE DIETRICH PETROLEUM BURNER.]
The regulation may be performed with the greatest ease. It is possible
to instantly vary, together or separately, the steam and the petroleum.
Under such circumstances, choking is not to be feared at the petroleum
orifice, where, according to experiment, the thickness of the substance
to be vaporized should not be less than 0.04 of an inch.
The petroleum might evidently be made to enter at A and the steam at
B; but one of the conclusions of the experiments cited is that the
performance is better when the jet of steam surrounds the petroleum. It
will be understood, in fact, that by this means not a particle of the
liquid can escape vaporization and, consequently, combustion.
Moreover, as the jet of petroleum is completely surrounded by steam its
flow can be increased within the widest limits, and this, in certain cases,
may prevent an obstruction without much diminishing the useful effect
of the burner.
The apparatus is easily and rapidly taken apart. It it is only necessary to
remove the nozzle, C, in order to partially clean it. It would even seem
that the cleaning might be done automatically by occasionally reversing
the flow of the steam and petroleum. However efficacious such a
method might prove, the apparatus as we have described it can be very
easily applied to any generator. Fig. 2 represents it as applied to the
front of a furnace provided with two doors. A metallic box, with two
compartments, is placed on one side of the furnace, and is provided
with two stuffing boxes that are capable of revolving around the steam
and petroleum pipes. The latter thus form the pivots of the hinge that
allows of the play of the vaporizers and piping.
[Illustration: FIG. 2--THE BURNER APPLIED TO THE FURNACE
OF A BOILER.]
It was in this way that Mr. Dietrich arranged his apparatus in an
experiment made upon a stationary boiler belonging to a Mr. Corpet.
The experiment was satisfactory and led to the adoption of the
arrangement shown in Fig. 3. The
Continue reading on your phone by scaning this QR Code
Tip: The current page has been bookmarked automatically. If you wish to continue reading later, just open the
Dertz Homepage, and click on the 'continue reading' link at the bottom of the page.